fair values in global accounting - ideal way or wrong track? articles and poems... · 2015. 1....
TRANSCRIPT
-
1
Fair values in global accounting - ideal way or wrong track? By Professor Rob Bryer, Warwick Business School, University of Warwick Coventry, UK.
Ludwig-Erhard-Foundation-Professorship Guest Lecture, University of
Bayreuth, 28th October 2004
‘Mirror, mirror, on the wall, which is the fairest value of all?’ [With
apologies to Jacob and Wilhelm Grimm.] Dear President, Dean, Ladies, Gentlemen, Professors, Doctors, Assistants and
Students,
Thank you very much for inviting me to be your Ludwig-Erhard-Foundation
Professor at Bayreuth University for 2004. I am greatly honoured, and glad to be here
again to see old friends and make new ones. I am here for two weeks to participate in
a seminar on international accounting with the advanced undergraduate accounting
students; to give this lecture; and to lecture on ‘issues in global accounting’.
‘Global accounting’ is my reason for being here - the real prospect that at least all the
listed companies of the major countries of the world will prepare their accounts using
the same rules. A spectacular step towards this goal was the EU’s decision in 2002
that all its listed groups must prepare their consolidated accounts using International
Accounting Standards or International Financial Reporting Standards (IAS or IFRS).
Following the EU’s decision, Australia, New Zealand, Russia and China, agreed to
implement IAS/IFRS. By 2005, 91 countries will either allow or require IAS, and it
looks likely that the US will join, allowing companies listed there to use IAS/IFRS
without reconciliation to US GAAP from 2007. We used to talk about ‘international
accounting’, but this is global accounting!
I have followed this historic development (that began in earnest the 1970s) through
my teaching and research for the last 15 years or so. Now, against all the expert
predictions, we are on the way to global accounting - at least, that is the destination.
To get there will require continued agreement by many countries, based on mutual
understanding. This is the main reason I am here today. I have come to meet German
scholars and others interested in discussing the direction of IAS/IFRS as Germany
makes major changes to its well-established system of accounting and taxation; to
offer a critique of the type of ‘Anglo-Saxon’ global accounting that is emerging under
US (and UK) influence; and to encourage you to join me in developing this critique!
I should first explain something about the title of the lecture; the rest will become
clear. Eighteenth century British merchants, industrial entrepreneurs and authors of
accounting books often talked of accounting as the ‘mirror’ that reflected reality and
allowed them to control their enterprises. Today, our image of accounting is not so
clear and, as Enron has reminded us, the images it provides users may not reflect
economic reality. Accounting is a large and complex subject, and is getting larger and
more complex by the day as the global economy and global society develops apace.
To face the challenges, we must go back to basics. Nothing is more basic in
accounting than what we mean by ‘value’, particularly the idea of ‘fair value’
accounting. US accounting scholars (such as William Paton and Andrew Littleton)
used this phrase in debates of the 1930s and 40s, and resurrected it in the 1960s in US
utility rate-setting cases. Today ‘fair value accounting’ it is on the lips of everyone
with a specialist interest in the subject. (Some evidence of this is that a Google search
produced about 5,360 hits on ‘historical cost accounting’, but 11,900 for ‘fair value
accounting’.)
‘Fair value accounting’ is the topic of my lecture today. I shall explain what ‘fair
-
2
value’ is; why the International Accounting Standards Board’s (IASB) predecessor
(the International Accounting Standards Committee (IASC)) made increasing use of
them; and why, many believe, the IASB wants a comprehensive fair value system.
Finally, I shall explain my view that the IASB’s approach to fair value accounting is
the wrong track to the future of global accounting.
I will try to convince you that ‘fair value accounting’ is not an arcane subject, fit only
for accounting professionals and academics, but raises important issues for all those
affected by the measurement of business performance (investors, government,
workers, customers, creditors) - in short, everyone. I will argue that the IASB’s
approach to fair value accounting redefines what we mean by ‘value’ in accounting
and, hence, what we mean by business success – what we mean by ‘profit’ – that
should be the subject of public discussion. My intention today is to provide non-
accountants with an introduction to some of the major issues.
At its broadest, ‘fair value’ simply means the market price - the value placed on a
product or service in a ‘fair’ market. However, as in reality there are potentially many
market prices, to decide which price is the ‘fair value’ we first need to be clear about
the fundamental aim of financial reporting to shareholders, creditors and other
outsiders. I will then ask what we mean by ‘fair values’, and show that aim we give
accounting determines which of the many ‘fair values’ we could use is (or are) the
‘fairest of them all’. Finally, I will draw some conclusions.
The primary aim of financial reporting: accountability or decision-relevance?
In the modern Anglo-Saxon world, large corporations relying heavily on equity
capital from large numbers of widely-diversified shareholders, and relatively little on
debt finance from banks, dominate economic activity. Strongly influenced by the
works of economists Irving Fisher and John Canning, during the late 1960s the US
accounting authorities concluded that, within the context of well-developed capital
markets, the aim of financial reporting should be to help equity investors decide
whether to buy, sell or hold shares. They called this aim ‘decision–relevance’, and
concluded that traditional aim of ‘accountability’ (explained below) was no longer
relevant.
We can challenge the historical premise of the FASB’s conclusion that growth in the
importance of stock markets justified changing the aim of accounting. Stock markets
have been important for much longer in the Anglo-Saxon world than the spectacular
flowering of the capital markets system in the US since the end of World War II
might suggest. Joint stock companies became the driving force of industrial
development in Britain from the end of the eighteenth century, and were supreme
from around 1850. Widely held joint stock companies became the dominant force in
North America only at the end of the nineteenth century, and they grew to spectacular
proportions during the early twentieth century. Well before then, however, British
judges, businessmen and accountants had forged the traditional principles that still
underlie the bulk of financial reporting practice, including US practice. There is, in
short, nothing in the history of capital markets to justify changing the aim of
accounting.
Throughout the history of capital markets, outside investors have demanded accounts
to make management ‘accountable’ to them for their capital and its profitable
employment – to give them ultimate control of management. How does accountability
do this? What we mean by ‘accountability’ in accounting comes from the two
meanings of the English word ‘account’. One meaning is the demand of a principal
that a subordinate agent produce an ‘account’ or reckoning of his or her performance.
The other meaning is that the principal then judges the account against a target and
-
3
punishes or rewards the subordinate accordingly. In modern management control
theory, we call accountability ‘results control’. Accounting therefore gives the
principal control of the agent by making him or her accountable for target results, for
example, a required return on capital. (The alternative is ‘action control’ - telling
managers what to do - but this is not possible for ill-informed outsiders.) Managers
know that the principal will judge an objective measure of their performance against a
target, and that the principal will punish or reward them on the outcome. This
motivates them to engage in what modern management control theory calls
‘feedforward control’, that is, planning ahead and taking corrective actions to achieve
targets (just as the prospect of examinations should motivate our students to work!).
I have said that the aim we choose for accounting will determine what we think is a
‘fair value’. My next questions, therefore, are: what is ‘fair value’, and how do the
different aims lead to different fair values? Finally, I will ask which aim is the most 6
important, and which, of the many alternative fair values that could stare back at the
users from the mirror of accounting, is the ‘fairest value of all’?
What are ‘fair’ values?
The English word ‘fair’, like many other English words, has two main meanings. It
either means ‘beautiful’, or it means ‘unbiased’. The meanings are quite distinct.
In Medieval English (Latin/Old French) a ‘fair’ was a periodical gathering of buyers
and sellers (usually on a holiday) where rules from a charter, statute or custom
governed the trading. The prices from such are market were ‘fair’ or unbiased because
they came from the application of agreed rules. In modern English, use of the word
‘fair’ often means unbiased (‘just’, ‘middling’, ‘average’).
By contrast, in Old English (Scandinavian/Gothic), ‘fair’ meant ‘beautiful to the eye’,
as it still does in modern English when we use it to mean ‘free from fault’;
‘favourable’; or to describe blond skin or hair.
These two English meanings of ‘fair’ are clearly different. It is one of the joys (and,
perhaps, one of the curses) of the English language, that we quite often have the same
word spelt in the same way having different meanings. (Other examples are, ‘right’
meaning legal or moral power, and ‘right’ meaning the opposite to left; ‘fine’
meaning good or alright, and ‘fine’ meaning a financial penalty; and, of course, the
word ‘account’ itself).
It certainly a curse on students of accounting that we now, in effect, have the two
meanings of the word ‘fair’ to describe accounting values - unbiased and beautiful –
that, we shall see, the authorities do not clearly distinguish!
To understand the meaning of English words such as ‘fair’ we must (as in every
language) understand the context of its use. In accounting in Britain, the traditional
(and legal) context for using the word ‘fair’ has, since the seventeenth century, always
been ‘true’ (or something that meant this, e.g., ‘proper’). Accounts must be ‘true and
fair’ - both together, and in the financial statements themselves, not in the supporting
notes. By contrast, as we shall see, the IASB only wants accounts to be ‘fair’, to be
‘beautiful’ as well as unbiased! The question for us is whether, in its obsessive search
for theoretical beauty, the IASB has lost touch with reality?
In the remainder of the lecture I will argue that accounting should not seek the IASB’s
theoretically beautiful ideal. That, indeed, accounting has its beauty, but it is the
practical beauty of a well-oiled, excellently designed machine, based on the best that
engineering science can offer. (The images it conjures up for me are of powerful
locomotives, steel plants, and German motor cars, not English roses or the Mona
Lisa!)
-
4
The IASB has made no official statement to this effect, but it is clear that it wants to
use more ‘fair values’ because it believes them to be the most theoretically perfect,
most beautiful, measures of value. Fair value is important in IASs 16, 18, 20, 21, and
IFRS 1, and is central to IASs 32, 39, 40 and 41 and IFRS 2, 3 and 4. The IASB
defines ‘fair value’ as “the amount for which an asset could be exchanged, or a
liability settled, between knowledgeable, willing parties in an arm’s length
transaction” [Glossary].
This definition does not help us very much. Knowledgeable, willing and otherwise
unrelated parties (‘at arm’s length’) could exchange an asset or settle a liability at a
variety of ‘market prices’. The question is which market prices are ‘fair’, that is, what
knowledge should a party to a transaction have to willingly exchange money for an
asset or a liability?
Consider the logical possibilities that exist in valuing a partly finished product (or a
service) for exchange at that point. (The possibilities are fewer for initial inputs or
finished products, but the issues are the same. We can think of a fixed asset as a partly
finished commodity; as a partly finished component of the final product). Its value
may vary along three dimensions: (i) when we value the product, (ii) the form in
which we value the product, and (iii) the market in which we value the product.
The values we can put on any product can refer to its past prices, to its current price,
or to its future price. The prices may depend on which market we are dealing with -
whether we are buying, when ‘entry’ prices are relevant, or selling when ‘exit’ prices
are relevant. Prices may also depend on whether we are valuing the initial inputs, the
asset in its current form, or the asset in its finished form less the costs to completion.
The value may also depend on whether we intend to sell the finished product, or to
use it. Leaving aside questions of using the initial inputs and the part-finished
commodity, or selling the initial inputs, we have 24 possible market values, although
we can easily dismiss the historical values as obviously ‘unfair’: Possible market values for a part-finished commodity
Form of asset
Value date, Market
Initial inputs
Present Form
Ultimate form
Disposal
Use
Past, entry
Historical costs of the inputs. The past cost for which the part- complete product could have been bought.
The past cost at which the finished product could have been bought, less the past costs to completion.
The past cost at which the stream of net benefits from a finished product could have been bought less
the past costs to completion.
Past, exit
Past selling prices of the inputs.
Past selling prices of the part- complete product.
Past selling prices at which the product could have been sold in its finished form, less the past costs to
completion.
Past selling prices of the stream of net benefits from the product in its finished form, less the past costs
of completion.
Current, entry Current prices of buying the inputs.
Current price of buying a part- complete product.
Current price of buying the finished product, less the current costs to completion.
Current price of buying the stream of net benefits from the product in its finished form less the current
cost to completion.
Current, exit
-
5
Current prices from selling the inputs.
Current prices from selling a part-complete product.
Current price from selling the finished product, less the current costs to completion.
Current price from selling the stream of net benefits from the finished product, less the current cost to
completion.
Future, entry Expected buying prices of the inputs.
Expected buying prices for a part- complete product.
Expected cost of buying the finished product, less the expected costs to completion
Expected cost of buying the stream of net benefits, less the expected costs to completion.
Future, exit
Expected selling prices of the inputs
Expected selling prices of a part- complete product.
Expected prices of selling the finished product, less the expected costs to completion.
Expected prices of selling the stream of net benefits from the finished product, less the expected costs
to completion
Historical cost is the fair value at the time of exchange because the buyer and seller
agreed this price. At that moment, the historical price is the current price. From that
moment onwards, however, the historical prices in the table are obviously unfair
because (except by chance) they are not prices at which knowledgeable and willing
parties would exchange now.
Most people would accept that to be ‘fair’ the value of a product should be the current
or future market price. Few would consider an offer to buy or sell at the past cost of
the inputs, or the past cost of buying or selling a part-finished commodity, or the past
cost of a finished commodity, a ‘fair’ basis for exchange. Few would consider past
selling prices ‘fair’ either, for whatever form of the commodity.
However, in theory, all the other prices in the table are ‘fair’ values at which informed
and willing buyers and sellers could exchange. In fact, if we follow the normal
economic assumptions about well-functioning markets, they should all produce
exactly the same price. (I leave aside the issues of the bid-ask spread, and whether we
should deduct transactions costs).
Assume that everyone earns the normal rate of profit, say, 25% on the cost of
production and distribution, that initial inputs include labour, and that the ‘costs to
completion’ include the normal rate of profit. Thus, in the above table, the current
costs and exit prices of the initial inputs should equal the cost of a part-completed
product. Valuing the part-completed product by reference to the final entry or exit
prices less the current costs to completion should give the same answer. Thus, for
example, assume that the selling price of a finished product to the consumer is £15,
the cost of production is £8, and the cost of distribution is £4, giving an overall profit
of £3. If the producer sold the finished product to a distributor the buying price would
be £10 (including £2 profit) and the distributor would expend a further £4 making a
profit of £1 by selling at £15. If the initial producer sold the commodity half-finished
when he or she has incurred costs of £4, the price to the buyer will be £5 including £1
profit. To this the second producer adds a further £4 costs to complete the product and
£1 for profit and sells it for £10 to a distributor, or spends another £4 on distribution
making another profit by selling at £15. In short: Current exit and entry prices in the part-complete, wholesale and retail markets in well-
functioning markets
Exit price for finished @ retail
Exit/entry prices for finished @ wholesale
Exit/entry prices for part-complete £5
cost of production =
£15
£10
-
6
profit = £1
cost of distribution =
£4
profit = £1
cost of production =
£4 profit = £1
£4 Time 0123
With appropriate changes to reflect expectations, future prices should in theory equal
current prices in all markets with appropriate adjustments for risk and time
preferences.
In short, according to economic theory we can only exclude historical values as
obviously ‘unfair’. From the remaining values we must choose which are ‘fair’ and
which are not (the table below indicates the traditional choice and the IASB’s choice,
which we explain later): POSSIBLE ‘FAIR’ VALUES FOR A PART-FINISHED COMMODITY
Value date/ Market
Past/entry
Past/exit
Current/entry
Current/exit
Future/entry
Future/exit
Form of asset
Present form
Unfair
Unfair
Initial inputs
Unfair
Unfair
Current reproduction cost replacement cost
Ultimate form
Use
Unfair
Unfair
Current economic value
Current selling price market value economic value
Disposal
Unfair
Unfair
Current market value
Current
Opportunity cost
Expected reproduction cost
Expected opportunity cost
Current
Current
Expected replacement cost
Expected market value
Expected economic value
Expected selling Price market value economic value
Expected
Expected = Traditional fair values
= IASB fair values
= Both
In reality, the prices in these different markets may be, and often are, very different.
For example, US experience in the used equipment market is that the exit market
-
7
prices plus transportation and installation costs do not equal the ‘cost of reproduction
new less depreciation’, the currently standard measure for US valuers (see, for
example, Alfred M. King, ‘New Rules for Fair V alue’, V aluation
Strategies, September/October, 2004, p.45.)
So, assuming for the present that prices exist in all these markets, and that they are
unbiased (but not necessarily the same), to choose between the alternatives we must
bring to bear our view of the primary aim of financial reporting. Do we want
accounting to hold managers accountable for the value of the capital they control, or
do we want it to help investors value their shares? What aim we choose will
determine which of the values in the table above we think is (or are) the ‘fairest
value(s) of all’.
Fair values for accountability
Following the traditional road, our overriding aim is to choose fair values that make
management’s accountable for the ‘capital’ they control, ‘the most important money’
that cycles around a business - sometimes for use; sometimes for disposal. The
accountant’s job is to make the cycles of capital visible to outsiders so they can hold
management accountable for the results. The figure below gives an outline of the total
cycle of capital through its different phases, during which it exists either in the form
of money or a claim to money (mainly as cash or debtors), or as useful things for the
business such as buildings, machines, inventories, etc., that is, as ‘non-monetary’
assets:
The enterprise operating cycle m
This figure says that outsiders put money (the capital) into a business (M) that
management spend on commodities (C), labour (L) and the means of production (mp)
that it puts into a production process (P). Out from production comes different
commodities (C’) that management, to continue in business, must sell for more
money Floating Capital
M′ Start here
C′ M Productive
Capital L
PC
mp
than they cost to produce (M’). If management makes a profit (m) it can distribute this
to investors, government, workers, etc., or invest it to expand the size of the capital
employed in the business.
The cycle of capital has two main phases, and here I shall use the old-fashioned
terminology to highlight the difference between them. In the first and last parts of the
cycle, accountants used to say that capital ‘floats’ on the market – either as capital
coming into a business (as cash or debtors) or as finished commodities coming out of
production or other assets available for sale. In between, capital goes through
production, is ‘productive’. To keep production going, management must replace the
non-monetary assets it consumes, and recover its monetary assets before it can
distribute any money as profit. The general rules in traditional accounting are,
therefore, that the fair value of non-monetary capital is its current replacement
(reproduction) cost (RC), and the fair value of monetary capital is its historical cost
(HC).
If special circumstances mean that the current replacement cost (buying price) of a
-
8
non-monetary asset is less than the current cost of identical assets in normal
circumstances, management must write down this asset to its ‘recoverable amount’,
the current replacement price given the special restrictions. Examples of special
restrictions requiring a write down to recoverable amount are state regulation of
production to below planned capacity, and loss of a major customer for a dedicated,
special-purpose plant. By definition, such write-downs should be rare events.
Similarly, if special circumstances or conditions prevent management from
recovering the historical cost of productive monetary assets, it must write these down
to their ‘recoverable amount’, the amount of the original cost that management can
recover under the special circumstances (e.g., bankruptcy of a long-term debtor).
For the same reason - accountability for the capital management actually controls -
traditional accountants will choose current exit prices as the fair value if this is less
than the current replacement cost of a non-monetary asset, and less than historical
costs of floating monetary assets. In other words, they use the very old ‘lower of cost
or market rule’. The table below summarizes the traditional fair value rules, and gives
some common examples of each category (explaining how the rules apply to items
such as investments in shares and financial instruments such as options and
derivatives, is beyond the scope of this lecture):
Traditional fair values
Productive Capital (assets in
Floating Capital (assets for
use)
disposal) Monetary
Monetary
Non-monetary
Non-monetary
Lower of
Lower of Historical Cost or Recoverable Amount.
Lower of
Lower of Historical Cost or Market (selling price).
Replacement
Replacement
Cost or
Cost or Market
Recoverable
(selling price).
Amount.
•Fixed assets • Loans • Leases • Mortgages • Finished • Debtors •Raw materials stocks
•Assets for sale •Work-in- progress
Replacement markets do not exist for some part-finished (or used), non-monetary
assets, or market imperfections may mean that the prices are biased. Nevertheless, the
primary markets for their inputs (labour and materials) do exist, and we can
reasonably assume that their prices are unbiased.
Economists see asymmetry in the traditional lower-of-cost-or-market rule that writes
current assets down to the market price when less than cost, but not up to market price
-
9
when this is greater than cost. Traditional accountants, however, see perfect symmetry
in accountability for capital controlled: when the market evidence is that management
has lost capital under its control, a write down holds them accountable. When the
market price rises above cost, traditional accountants disregard this because
management has yet to control (that is, to realise) this increase and so cannot take
credit for it in the accounts.
The lower of cost or market rule EXIT PRICE
COST EXIT PRICE
UNREALISED GAIN CONTROLLED CAPITAL LOSS OF CONTROLLED CAPITAL
RECOVERABLE CAPITAL
To hold management accountable, accounts must not only be ‘fair’ (unbiased), but
‘true and fair’ - that is, objective and unbiased. For decision making, by contrast,
accounts need only be ‘fair’, and this is why, I think, it sends us down completely the
wrong track.
Fair values for decision-usefulness
If we follow the IASB’s decision-usefulness objective for accounting, we will ideally
choose as the fairest (most beautiful) of ‘fair values’ the current or expected selling
price, or the expected market value or economic value. The reason is that all are
measures of the expected present value of the product which, according to the FASB
and its supporters, is the value with 100% relevance to investment decision-making.
Alternatively, we can limit the choice of fair values to current exit prices, current
economic value, or current market value. Current economic values and current market
values constrain estimates of the present value to the seller to current prices and costs
(ruling out management’s use of expected prices and costs).
The IASB does not give any general guidance about selecting fair values, only
specific rules in specific standards. The nearest to its current thinking is probably
current exit prices and expected economic value using exit prices, as it is for the
FASB and for Sir David Tweedie whose theoretical preferences derive in part from
the works of Raymond Chambers, a well-known advocate of exit values for financial
reporting. The FASB recently published draft rules on Fair Value Measurement
where it makes its preference for market prices clear by insisting that only if unbiased
market prices do not exist can valuers even think of using the equally important cost
or income approaches to valuation. Level
Fair value Market I
Quoted prices for identical assets or liabilities in reference markets whenever that
information is available.
Market II
Quoted prices for similar assets or liabilities in active markets, adjusted, as appropriate for
differences, whenever that information is available.
Multiple Valuation Techniques
If quoted prices...are not available, or if differences ...are not objectively determinable, fair
value shall be estimated using... the market approach, income approach, and cost approach
whenever the information necessary...is available without undue cost and effort’
Although quoted prices are objective, requiring management to use ‘adjusted’ prices
of ‘similar’ assets, or the ‘income approach’ (that forecasts and capitalises cash flows)
if it thinks the expense of gathering the information is not ‘undue’, shows that the
FASB’s definitions leave the door wide open to subjectivity.
-
10
Furthermore, whereas current costs are objective because they measure the value of
inputs actually consumed at current prices, although current exit prices are objective,
they are not objective measures of management’s performance as only by chance do
they measure the exit price at the date of sale. Market or economic values based on
current or future exit prices add further subjectivity as we do not objectively know the
future costs to completion. As the FASB admits in its exposure draft,
“The objective of fair value measurement is to estimate an exchange price for the
asset or liability being measured in the absence of an actual transaction for that asset
or liability. Thus, the estimate is determined by reference to a current hypothetical
transaction between willing parties” (Proposed Statement of Financial Reporting
Standards, Fair Value Measurements, September 2004, para.5, emphases added).
Exit prices are necessary for decision ‘relevance’ or economic ‘beauty’, but they are
‘hypothetical’ and, therefore, inherently subjective as measures of management’s
performance. Where active markets do not exist (and sometimes when they do),
management can bias prices (as Enron has reminded us).
Accountants usually call expected exit value ‘net realisable value’ (NRV). Net
realisable value is subjective because it asks management to estimate future inputs at
current prices. By contrast, to construct a replacement cost from the initial inputs we
only ask management to put current prices on past (or established, technically
improved) inputs. (For example, whereas there is no market for the used wiring and
piping in a chemical plant - often a large part of the total value - we know the current
prices of replacing it new, and can estimate how much economic life it has left.)
If we believe that the aim of financial reporting should be decision-relevance, we will
report all increases or decreases in the market prices of all assets and liabilities as
profits or losses. That is, we will account for increases in market values as a surplus
that the government could tax (and account for decrease as a deficiency that could
relieve tax), be divided (or restrict dividends) or reinvested (or restrict investment).
However, many (if not most) of these prices changes may be temporary fluctuations,
and management has realised none of them.
In contrast to the IASB’s approach to fair value accounting, traditional accountants
account for all increases and decreases in current replacement cost of non-monetary
assets as capital maintenance adjustments (CMA). That is, adjustments to changes in
the recoverable value of the entity’s capital that are beyond management’s control.
We can see the potentially dramatic different results from the traditional and decision-
relevance approaches to accounting for profit and loss in the following diagram: Traditional versus decision-relevance fair value accounting for profits and losses
Prices
HC = RC = NRV
NRV RC RC
Exit price
NRV
0123 N Time
Profit?
CMA+Loss
CMA or Profit?
Loss
Loss?
The diagram shows that we bought an asset at time 0 at which time the price equalled
the RC and NRV. During the time 0-1 the replacement cost increases above HC and
the NRV even more. At the end of the first period traditional accountants write up the
asset to its RC as a capital maintenance adjustment (CMA), whereas decision-relevant
-
11
accountants write the asset up to NRV and book a profit (NRV – HC). At the end of
period 1 (beginning of period 2) the RC increases again, stays there until the end of 2
when the exit price falls, and the NRV even more. The traditional accountant first
writes up the asset by a CMA for the increase in RC and, at the end of period 2, books
a loss for the fall in exit price to below RC. The decision-relevant accountant, by
contrast, writes down the asset to its lower NRV, booking a bigger loss. From this
simple illustration, where the decision-relevant profit and loss exaggerates the swings
in traditional profit and loss, we can easily appreciate the often stated fear that the
IASB’s fair value accounting will cause increased volatility in profit that is
completely spurious from the accountability viewpoint.
The IASB claims that decision-relevance subsumes ‘accountability’. That investors
can hold management accountable by selling shares (punishing it by increasing the
cost of capital) or firing people if the share price (or ‘fair value’ of the firm in the
accounts, for the IASB, an approximation for market value) falls, or buying shares
and acceding to higher salaries and pensions, etc., if it (or the ‘fair value’) rises. This
view, however, overlooks that we cannot objectively attribute fluctuations in share
prices solely to management actions because many of them reflect events (e.g.,
interest rate changes, technical changes) beyond its control, or result from
speculations about future events. The same is true for decision-relevant profits and
losses, fluctuations in the ‘fair value’ of the firm, as we have seen.
Going for decision-relevance inevitably means subjectivity, and this means giving up
accountability because it means managers taking the credit for unrealised gains and
being punished for unrealised losses, that will never arise. It is easy to understand
why, therefore, British, American, German, French, Japanese (etc.) accountants and
commentators have heavily criticised the subjectivity of decision-usefulness
accounting, and they continue to complain. Several commentators have noted that
none of Enron’s tricks with special purpose entities would have been possible if US
GAAP had forbidden mark-to-market accounting.
In response to early vociferous concerns by British accountants about subjectivity,
proponents of current value accounting (notably, Sir David Tweedie, who wrote the
ASB’s Statement of Principles) have often (somewhat reluctantly) limited
management’s choice of fair values with the ‘value-to-the-owner’, or ‘deprival value’
rules, first developed for insurance companies: Deprival values as fair values
If fire, for example, deprived the owner of his or her factory, an insurance company
would compensate the owner by replacing the factory because the owner can then use
the factory to earn whatever economic value it has. If the economic value of the Deprival value
Lower of
Replacement cost Recoverable amount
Higher of
Net realisable
value
Economic value
factory is less than its replacement cost, the owner would either sell it, or use it and
not replace it, and would be compensated for loss by the higher of these values.
Deprival value rules are part of the British Statement of Principles and are implicit in
IAS 36: Impairment of Assets, and the IASB will probably include them when it
revises its conceptual framework.
Deprival value rules puts replacement cost as the highest fair value, just like
traditional accounting, but the other rules are very different. Compared to traditional
-
12
fair values, deprival value rules introduce subjectivity by requiring write downs of
current assets to NRV rather than to exit price, and by requiring assets to be written
down to economic value rather than the traditional measure of ‘recoverable amount’
(that is, to restricted replacement cost).
Deprival value rules also introduce subjectivity by requiring entities to carry all assets
and liabilities at current market or economic values, including its long-term debt. As
David Damant, a leading advocate of decision-usefulness accounting, rightly says,
“the revaluation of an enterprise’s own long-term debt is something which the large
majority of users of the traditional accounts find very difficult to accept” (Financial
Times, 20 June 2002, emphasis added).
Many preparers and users of accounts find the prospect of valuing debt at fair value
unacceptable because the consequences are paradoxical. The IASB admitted that one
of the concerns of bankers, securities regulators and insurers over IAS39 (the rules for
financial instruments, such as derivatives) was that “if an entity applied the fair value
option to financial liabilities, it might result in the entity recognising gains or losses in
profit or loss for the changes in its own credit-worthiness” (International Accounting
Standard IAS 39, Financial Instruments: Recognition and Measurement (London:
IASB, revised 2003, Background, para.3(c)). This absurd result may “explain...why it has proved so difficult to produce rules that more than a handful of theoreticians will
accept” (Robert Bruce, Financial Times, 20th June 2002).
We can, of course, explain how much of the ‘profit’ is down to falling credit-
worthiness - “the amount of the change in the fair value that is not attributable to
changes in a benchmark interest rate” (IASB, 2003, para.BC4). However, this does
not explain why any fall in the current value of debt is ‘profit’ in the first place. We
can, of course, argue that there is no profit if we write down the assets to offset it, but
there can be no automatic need to write off assets because the market value of a firm’s
debt falls, and we would still have to explain the ‘profit’ to equity, albeit one offset by
a loss.
If we use fair value accounting and management uses derivatives or otherwise hedges
assets and liabilities to eliminate risks, we get spurious volatility in the components of
earnings that cancel out in the profit and loss account with no net effect. However, if
we do not mark both sides of a hedge to market – do not use comprehensive fair
values for all financial instruments - we will also get spurious volatility in total
earnings (or in equity for some hedges according to the rules of IAS39).
A paradoxical example of spurious volatility is the issue of fairly valuing a bank’s
demand deposits. Is the fair value of a liability repayable on demand its face value (as
a traditional accountant would argue), or is it their present value, that is, the deposit
less the returns on it the banker expects before the customer withdraws it? One would
think the IASB would naturally choose the latter, but the unacceptable consequence
would be the recognition of a profit from merely acquiring a deposit!
Imagine, then, the frustration of the European (particularly French) banks in being
unable to use fair value hedge accounting (and account for offsetting gains and losses
on the derivative and the hedged item) for their hedges of the interest rate risk on
demand deposits. Instead, the IASB insists the banks must account for demand
deposits at their historical proceeds which, according to the rules of IAS39, means
they must account for the derivative at fluctuating fair market values in their equity
accounts as a ‘cash flow hedge’. This could induce spurious volatility in the banks’
equity, with potentially serious consequences (for example, they could need to raise
extra equity to protect bond covenants) in addition to confusing shareholders and
others!
-
13
In the face of this nonsense, on the 1st October the EU decided to give member
countries the option of limiting the application of IAS39’s hedging rules. On 1st
October the Accounting Regulatory Committee of the European Commission voted to
remove the IAS fair value option as it applies to liabilities, and to facilitate the use of
fair value hedge accounting for interest rate hedges of core deposits!
Furthermore, in practice (and contrary to common belief), many companies have their
derivative financial instruments custom-made for their particular circumstances. By
definition, ‘active’ and ‘homogenous’ markets do not exist for these derivatives.
Therefore, management must value them using models that are sensitive to small
changes in assumptions. We cannot, therefore, assume that these valuations are
unbiased.
To sum up:
Partial use of fair values partially undermines management’s accountability for
capital. Even if we apply fair values to only those assets and liabilities for which
unbiased market prices exist, fair value accounting remains subjective and it increases
the reported volatility of earnings.
Full use of the IASB’s fair values will seriously undermine management’s
accountability. If we apply fair values fully, we open the door wide to subjectivity and
may well introduce bias. A comprehensive fair value accounting system would allow
management to ‘mark-to-market’ or ‘mark-to-model’ all of its assets and liabilities.
This would allow it to take credit for unrealised gains, and require asset ‘impairments’
to be written down to economic value as losses (and reversals as profits). When we
value debt at market value we get a result that is counter-intuitive.
Is compromise possible?
Many broadminded accounting scholars (and those tired of war) think that we need
not choose between accountability and decision-relevance - that we can have both the
tiresome ‘beast’ of accountability and the ‘beauty’ of decision-relevance (to mix my
fairy tale metaphors!).
In my view, we can have both, but not in the same financial statement. My fear is that
if we mix decision-relevant fair values with fair values for accountability, we risk
confusing the users and undermining accounting’s distinct accountability function.
We cannot trade-off a bit less accountability for a bit more decision-relevance
because they are mutually exclusive systems with mutually exclusive functions.
We can, of course, ask management for a separate ‘valuation statement’ (its estimate
of the net present value of the firm), but we must not mix this with ‘the accounts’,
whose accountability function is distinct, and must be kept distinct. The IASB says it
believes in the rationality of markets. Why not, therefore, produce two reports and let
the market of users and commentators decide which is most important?
Apart from the expense, it seems very unlikely that the IASB would ever agree to two
statements as it would mean recognising objective accountability as a primary aim of
accounting. This would mean a wholesale revision of its US-inspired conceptual
framework, in which many of its incumbents have invested their intellectual lives.
Robert H. Hertz, for example, chairman of the FASB, educated at the University of
Manchester, still considers himself a Hicksian economist and, therefore, believes that
accounting should define “income as changes in wealth” which Sir John Hicks
famously argued we should define as expected present value. Robert H. Hertz, not
surprisingly, fervently believes that “fair value is the most relevant measurement
attribute” CFO Magazine, February 01, 2003). Trapped in a lover’s trance, “the
FASB’s...cavalier approach to verifiability is troubling” (Ross L. Watts,
-
14
‘Conservatism in Accounting Part I: Explanations and Implications’, Accounting
Horizons, Vol. 17, No.3, 2003, pp.207-221), but wholly understandable.
True believers such as Hertz do not even recognise the aim of accountability. To wake
them from their sleep, we must confront them with the choice they are implicitly
making for us in trying to live out their dream of accounting in an ideal economic
world. We must tell them that this dream is becoming a nightmare; that although their
aim of decision-relevance offers a beautiful theoretical perspective, in the cold light of
day it is often ugly as unbiased markets for assets and liabilities do not exist, and the
‘prices’ that emerge from valuation models may be biased. To rouse them from their
dreams, we must convince the true believers that the real world demands the practical
beauty of accountability – that, in reality, economic ‘fair value’ is not the ‘fairest of
them all’. In short, the real choice is not between historical cost and economic fair
value accounting. In reality, we must choose between decision-relevant accounting
that is ugly in practice, and practically beautiful traditional accounting. Here is the
real choice as I see it:
‘MIRROR, MIRROR...’: THE REAL CHOICE
Conclusions
My overall conclusion is clearly that the IASB’s approach to fair value accounting is
the wrong track for global accounting because it undermines accountability. The
IASB’s approach could, therefore, impede the development of capital markets, or
even weaken them by undermining investors’ confidence, and it could impede or
weaken business accountability to government and labour.
The EU and the member governments on continental Europe have worked hard over
the last 25 years or so years to create an enduring ‘equity culture’. The IASB’s fair
value accounting could seriously undermine this effort. I agree with Sir David
Tweedie that
“After the Enron and WorldCom scandals...people realised how important accounting really is. Accounting is the bedrock of capitalist society, because if you can’t trust the
numbers, people won’t invest” (Accountancy, January 2004, p.56).
UNBIASED
BIASED
OBJECTIVE
Theoretically ugly historical cost accounting
Practically beautiful
traditional accounting
SUBJECTIVE
Theoretically beautiful decision-usefulness accounting
Practically ugly decision-
usefulness accounting
It is also true that if government and workers do not trust the numbers, they will not
invest in businesses – whether through tax relief for losses, or restraining wage
demands.
However, I disagree profoundly with Sir David’s view that to ‘tell-it-as-it-is’, “the
market is where reality is and that is why a lot of people don’t like it” (Accountancy,
January 2004, p.56). I agree that ‘the market is where [one] reality is’, but which
market is an important question (he means exit prices), and management and its
accountability is also ‘where reality is’.
I think that what would really make investors lose confidence in investing is if they
believe that management is not fully accountable for their capital that it controls.
-
15
If we believe that the IASB should not ignore accountability, what can we do? Before
we get too depressed about the formidable momentum behind the decision-relevance
paradigm, we should remember that securing the full co-operation of the EU is
essential to the survival and success of the IASB. Germany is a powerful member of
the EU that rightly takes its accounting seriously because, I think, it recognises that
logical rules for business are essential elements in the beneficial regulation of social
life. Fair value accounting threatens German accounting and, therefore, the German
way of life with its traditional emphasis on social accountability.
Before we also get too depressed about the apparent hegemony of a US that does not
listen, we should also note that, after a 30 year trend towards ‘fair value’ accounting,
‘Enronitis’ has caused Americans to seriously question their often self-confidently
asserted belief that ‘US accounts are the best’. As Watts says,
“the Enron case demonstrates...[that] the FASB appears to favor mark-to- market
accounting without insuring verifiability of the market estimates.... The FASB can ill
afford more scandals of the Enron variety in which ‘generally accepted’ unverifiable
values played a role” (op cit, p.218).
The same is true for the IASB which is following the FASB “down a path that many
before them have feared to tread, and with good reason” (op cit, p.219). I agree with
Watts that this path leads to a world where “net asset values and earnings are subject
to more manipulation and, accordingly are poorer measures of worth and
performance” (ibid) – to a world with less accountability.
Given its strong intellectual traditions in accounting, Germany ought to play a leading
role in conducting an effective European critique of the US and IASB’s ‘fair value’
accounting typified by IAS39. It was, therefore, somewhat disappointing that
Germany abstained in the June 2004 Accounting Regulation Committee vote on
IAS39!
I hope this short lecture has helped you to understand some of the issues involved in
an important public debate about ‘fair values’ in accounting that should be taking
place, but is not. Encouraging an effective European critique of the IASB’s approach
is an important reason for my visit, and for enthusiastically accepting your gracious
offer of the Ludwig-Erhard-Foundation-Professorship for this semester. Thank you
for listening.